Do No-Shows Really Cost US Healthcare $150 Billion a Year? The Source, Checked (2026)

The claim that patient no-shows cost US healthcare $150 billion a year has no published method behind it. The earliest dated appearance we could find is a February 2013 Pittsburgh Post-Gazette article that credits it only to “one account”, and the best-known version comes from the vendor SCI Solutions, in November 2016 press coverage of its report and a 2017 commentary by its Chief Marketing Officer, which presents the figure as following from about $200 per empty 60-minute slot and no-show rates “as high as 30%”, with no source and no formula. We found no MGMA or US government study that produces it, and measured no-show rates (an MGMA median of 5% to 7% for medical practices, about 23% pooled across 105 studies) give far smaller totals with the same $200 per slot.

The vendor's underlying point, that no-shows are costly, is fair. The number is the problem. It now sits in vendor blogs, university news and seven peer-reviewed articles we read (2023 to 2026), credited in turn to MGMA, to Health Management Technology, to Kyruus Health and to papers that do not contain it. This page follows every link we could open, shows the arithmetic, lists what we checked and did not find, and ends with the figure a practice can honestly use: its own. It is the same source audit we ran on the lead response time studies and on the 9.3 hours a week spent searching for information.

Where does the $150 billion no-show figure come from?

The earliest dated appearance we found is in a newspaper, not a study. On 24 February 2013, a Pittsburgh Post-Gazette article by Bill Toland, headlined “No-shows cost health care system billions”, said: “Such inefficiency, by one account, costs the U.S. health-care system more than $150 billion a year” (Pittsburgh Post-Gazette, 2013). The article does not say whose account it was. That is the earliest we found, not proof that nothing older exists.

The best-known version comes from a vendor. On 17 November 2016, Healthcare Finance News reported that “missed and open appointments cost the healthcare industry $150 billion annually, according to a new vendor report that recommends automating the process.” The report, from SCI Solutions, said “providers have no-show rates between five and 30 percent nationwide” and that “each 60-minute open or no-show slot typically costs physicians $200” (Healthcare Finance News, 2016). The article gives no sample and no method for the report, and we could not find the report itself.

Five months later, on 26 April 2017, SCI Solutions' Chief Marketing Officer, Jamie Gier, published the version that several journal papers cite as “Gier 2017”. Healthcare Innovation labels it “Commentary” (Healthcare Innovation, 2017). The three sentences that carry the claim:

There is no hyperlink or citation on the $150 billion sentence, no formula showing how $200 and 30% become $150 billion, and no mention of MGMA. “You can see how that figure is reached” is an invitation to do the arithmetic, so we do it below.

Which sources are behind the no-show numbers, and how reliable is each?

Every source in this audit that carries a number, with what it measured and how far it can be trusted. The first four rows are where the $150 billion lives. The rest are the measured data on no-show rates and costs.

How was the $150 billion calculated, and does the math hold up?

The commentary gives two inputs and no formula, so the only thing to test is what those inputs imply. What follows is our illustration of the vendor's formula, not a new national estimate.

At $200 per missed slot, $150 billion is 150,000,000,000 divided by 200: 750 million missed appointments a year. For scale, US office-based physicians had about 1.0 billion visits in 2019, according to the CDC's National Ambulatory Medical Care Survey. If those 1.0 billion are the visits patients kept, the missed appointments at a no-show rate r are 1.0 billion × r / (1 - r). That is the reading most generous to the vendor, because it adds the missed appointments on top of the visits that happened.

Even the vendor's own top rate of 30%, applied to every office visit in the country, gives about $86 billion. Reaching 750 million missed slots on top of 1.0 billion kept visits takes a no-show rate of about 43% (750 out of 1,750 booked). In the 105-study review, only the highest continent average reaches that level: Africa's 43.0%, against 13.2% for Oceania at the low end (Dantas et al., 2018). With MGMA's measured medians for US medical practices, the same formula gives $10.5 billion to $15 billion.

Did MGMA, Health Management Technology or Kyruus publish it?

MGMA

This is the most common attribution, and the one we checked most closely. A search of mgma.com for “150 billion” (29 October 2026) returns one document: a 2021 paper by a practice executive, “submitted for partial fulfillment of the requirements of Fellowship in the American College of Medical Practice Executives” and published in MGMA's Fellowship Papers section (MGMA Fellowship Papers, 2021). It says: “It is estimated that no shows cost the health care industry $150 billion each year.” There is no in-text citation on that sentence. The paper's works cited do list the February 2013 Pittsburgh Post-Gazette article that carries the figure, credited there only to “one account”. They list no SCI Solutions, Health Management Technology or national cost study, and the paper contains no “$200”. It is one member's paper, not MGMA research.

The same paper quotes MGMA's actual data: according to the MGMA 2016 Practice Operations Report, “most specialties have a median of 5 percent patient no shows, with neurology and general surgery being higher, at 8 percent.”

The MGMA pages that blogs link for “$150 billion (MGMA)” do not contain it. The August 2022 MGMA Stat article reports that “the median patient no-show rate for medical practices jumped by 40%, from 5% in 2018 to 7% in 2019” and that 49% of medical groups said their no-show rates had increased since 2021; it has neither “150 billion” nor “$200” (MGMA Stat, 2022). The August 2024 MGMA Stat poll, which a 2024 vendor blog links for both “$150 billion” and “10% to 30%” no-show rates, reports that “nearly four out of 10 medical groups report their no-show rates were on the rise this year” from 303 responses, and gives no no-show percentages at all (MGMA Stat, 2024).

We found no MGMA study, poll or data page that produces the figure. MGMA's paywalled DataDive reports were not read, so the accurate statement is “we found no MGMA study”, not “MGMA never said it”.

Health Management Technology

A January 2019 article on CustomerThink by a software vendor's president says: “Missed appointments cost the US healthcare system $150 billion annually, according to Health Management Technology,” linking an HMT address (CustomerThink, 2019). In the Internet Archive's 2019 captures, that HMT address redirects to the 2017 SCI Solutions commentary at Healthcare Innovation, and healthmgttech.com now redirects to Healthcare Innovation's homepage. The “HMT study” is the vendor commentary. An undated vendor page goes one step further and calls it “a study conducted by Health management Technology”.

Kyruus Health

One healthcare software vendor's blog says “Kyruus Health says no-shows cause about $150 billion in losses every year in healthcare,” with no link. Two searches for Kyruus and “$150 billion” found nothing behind it. This is a search-based finding, so it is stated at medium confidence: we found no Kyruus publication with the figure.

How did the $150 billion figure spread, all the way into journals?

In date order, every appearance we read, with the source each one gives. Vendor blogs are described rather than named, because they are evidence of the drift, not sources.

In the peer-reviewed articles we read, all from 2023 to 2026, the figure usually sits in the introduction or background, where it sets the scene for a study about something else. The studies themselves are about other questions; what matters here is what each reference contains.

• Citing the 2017 vendor commentary directly: Frontiers in Health Services (2023) writes “Data has shown that missed appointments cost the US health care system more than $150 billion a year and individual physicians an average of $200 per unused time slot”, citing the commentary and a 2013 simulation study by Berg and colleagues; Berg's abstract reports a daily loss, not a per-slot $200. Bioengineering (2025) and the Journal of Patient Experience (2026) also cite the commentary.
• Citing the MGMA fellowship paper as a study: the Journal of Public Health Policy (2026) writes that “one study found” no-shows cost more than $150 billion a year and physicians “an average of $200 per unused time slot”, citing the 2021 fellowship paper. That paper is not a study, gives no in-text source for the figure and has no $200.
• Credited to a paper that does not contain it: the International Dental Journal (2026) credits “$196 per patient and a total of US$150 billion a year” to Kheirkhah and colleagues (2016). The Kheirkhah paper has no “billion” anywhere in its full text; its $196 is the cost of a no-show at one Veterans Affairs medical centre. Cardiology Research (2023) writes that the no-show rate “is as high as 30%, constituting about $150 billion in healthcare costs annually”, citing a 2021 heart-failure telehealth paper by Sammour and colleagues whose full text contains no “billion” and no “150”.
• Called “an earlier study”: BMC Health Services Research (2023) cites a literature review of no-show behaviour, not a cost study.

Outside journals, a 2020 University of Arizona engineering news article repeats it with no source, and a 2025 vendor blog repeats it five times with no source while noting that “the $150 billion figure gets quoted a lot.”

Our own site is in this trail. Two earlier posts here, our guide to AI appointment scheduling and our healthcare AI adoption statistics, repeated the MGMA attribution and linked the 2022 MGMA Stat page, which does not contain it. Where they differ from this audit, this audit is the checked version.

What did we check and not find?

Absences are stated in the words below and no stronger. “We found no” is a claim about what we searched, not proof that nothing exists. All checks were made on 29 October 2026.

What is the real average patient no-show rate?

It depends on the setting, so the honest answer is a range with its sources, not one national number.

• Worldwide, across studies: a 2018 systematic review retrieved 727 records and selected 105 for analysis; “the average no-show rate is of the order of 23%, being highest in the African continent (43.0%) and lowest in Oceania (13.2%).” The most commonly reported predictors were long lead time and a history of earlier no-shows (Dantas et al., Health Policy, 2018). The abstract gives no North America figure.
• US medical practices: MGMA's benchmark median was 5% in 2018 and 7% in 2019 (MGMA Stat, 2022). MGMA's 2016 data, as quoted in the fellowship paper, put most specialties at 5% and neurology and general surgery at 8%.
• One large VA medical centre: a mean of 18.8% across its 10 main clinics; across all its clinics in 2008, 146,358 no-shows in 1,028,300 appointments (14.2%) (Kheirkhah et al., 2016).
• Single settings: 18% in a simulated endoscopy suite built on historical attendance (Berg et al., 2013) and 21.2% at an academic adolescent clinic in Michigan (Triemstra and Lowery, 2018).

The 5% to 7% and the 23% are not directly comparable: one is a median across US medical practices, the other an average across published studies from many countries and specialties.

Where does “30%” come from? The vendor commentary gives no source for it. The best-known study figure near it counts something broader: at one US family practice residency clinic, “no-shows and cancellations represented 31.1% of scheduled appointments” (Moore et al., 2001). We are not saying the vendor took its figure from that paper, only that the most-quoted “about 30%” study number counts no-shows and cancellations together.

For rates in other settings, see our dental no-show statistics, veterinary no-show statistics and restaurant no-show statistics.

How much does a no-show cost a practice?

The single-clinic studies are more useful than any national figure, because they show their method, and they show why one dollar figure per slot misleads.

• Billed is not paid. At the Michigan adolescent clinic, missed appointments cost an estimated $170,100 a year in billing but $51,289 in reimbursement, roughly a threefold difference (Triemstra and Lowery, 2018).
• Empty slots get partly refilled. At the residency clinic, “walk-in patients replaced 61.0% of failed appointments but only 42.4% of the time blocked for those appointments,” and the projected yearly revenue shortfall was 3% to 14% of clinic income (Moore et al., 2001).
• Cost is not lost revenue. Kheirkhah's $196 per no-show is the VA's cost per encounter, in 2008 dollars, at one health system; the marginal cost of no-shows in its 10 main clinics was $14.58 million in fiscal 2008 (Kheirkhah et al., 2016).
• Schedule policy changes the loss. In a simulated endoscopy suite booking 24 patients a day, an 18% no-show rate cost $725.42 a day in expected net gain (16.4%); the interventions tested cut the loss by $166.61 to $463.09, and overbooking 9 extra patients a day removed it (Berg et al., 2013).
• An audit that shows its arithmetic. The VA Inspector General projected that about 4.9 million (18%) of 26.5 million scheduled appointments in VHA's 10 performance-measure clinics went unused in fiscal 2008, counting no-shows, cancellations and unscheduled slots, and that reducing them would let VHA make better use of resources valued at about $76 million a year (VA OIG, 2008). That is agency level; we found no US government estimate of the national cost of no-shows.

What reduces no-shows?

The evidence on reminder and confirmation calls has its own page: how AI receptionists reduce no-shows. Four findings from the sources above point the same way:

• Lead time matters. Long lead time was among the most commonly reported predictors in the 105-study review, and non-attendance rose with lead days at the Michigan clinic (Dantas et al.; Triemstra and Lowery).
• A phone reminder system moved the rate only a little. At the VA medical centre, the no-show rate “remained at a high level despite its reduction by a centralized phone reminder (from 16.3% down to 15.8%)” (Kheirkhah et al.).
• Refilled slots recover money. Walk-ins replaced 61.0% of failed appointments at the residency clinic, and overbooking removed the loss in the endoscopy simulation. A cancellation heard about in advance gives the team a slot to refill instead of an empty one (Moore et al.; Berg et al.).
• Common levers are still unused. Fewer than 20% of single-specialty practices charged a no-show fee in 2019 (MGMA, 2022), and in an earlier MGMA poll, reported on MGMA's August 2024 Stat page about the 303-response poll, only 15% of medical groups reported using predictive analytics to improve no-show rates or patient scheduling (MGMA, 2024).

That is the job a two-way confirmation call does. The patient confirms, cancels or moves the appointment on the call, and a cancellation two days out becomes a slot the team can offer to someone on the waiting list. An AI voice agent can make those calls for a clinic's whole next-day list, the changes land in the booking system, and anything clinical goes to the team, so the front desk keeps its time for the patients in front of it. See how this works for medical clinics and dental clinics, and how AI answering works for medical practices. Booking work outside healthcare, such as sales meetings, has its own page: AI appointment setting.

If you want to see what confirmation calls would look like on your own schedule, book a call with us.